HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 22, 16758-16766, June 2, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/22/16758    most recent M000121200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Liu, D. Articles by Hew, C. L. Search for Related Content PubMed PubMed Citation Articles by Liu, D. Articles by Hew, C. L. Social Bookmarking                      What's this?

A Zebrafish Ftz-F1 (Fushi Tarazu Factor 1) Homologue Requires Multiple Subdomains in the D and E Regions for Its Transcriptional Activity^*

From the ^a Division of Structural Biology and Biochemistry, Hospital for Sick Children, Toronto, Ontario M5G 1L5, Canada, the ^b Departments of Biochemistry and Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario M5G 1L5, Canada, the ^e Department of Biology, ^h Center for Ligand and Transcription, and Hormone Research Center, Chonnam National University, Kwangju 500-757, Korea, and the ⁱ Department of Biological Sciences and Tropical Marine Sciences Institute, National University of Singapore, 119260 Singapore

Received for publication, January 6, 2000, and in revised form, February 23, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A zebrafish Ftz-F1 homologue, zFF1A (zebrafish Ff1a or Nr5a2, a member of nuclear receptor superfamily) and its C-terminally truncated variant (zFF1B) were previously identified. Due to lack of the identity box (I-box) and activation function 2 (AF-2) domain, zFF1B lacks transactivation function and fails to synergize with estrogen receptor (ER) in regulating promoters. It was speculated that the I-box might be involved in the zFF1A/ER interaction. In the present study, the function of the I-box was examined. In the absence of the I-box or with an altered heptad 9, the AF-2 of zFF1A was not functional, either in the presence or absence of ER. The GST pull-down assay showed that zFF1A and its mutants exerted similar physical contacts with ER-LBD, suggesting that the "dimerization" domain (I-box) is essential for the transcriptional activity of zFF1A. Moreover, nuclear receptor coactivator selectively activated zFF1 with the I-box but exerted no effect on zFF1B, indicating that the I-box is able to interact with the coactivators. By deletion study and analysis of the identified domains in GAL4-DNA binding domain, other regions of zFF1A critical for its AF were also delineated. Consistent with the mutation analysis, AF-2 was active only in the presence of the I-box. We also identified a novel AF domain (AF-3) located in the hinge region (amino acids 155-267), although the activity of AF-3 was inhibited by its flanking region. We suggest that the D and E regions of zFF1A possess both positive and negative transactivation functions, and interdomain "cross-talk" may confer the full transcriptional activity of the protein.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The nuclear receptor (NR)¹ superfamily comprises a large set of ligand-regulated transcription factors, whose modular structures provide functional regions responsible for their activities (1). In general, they contain an amino-terminal ligand-independent activation domain (region A/B, or activation function 1 (AF-1)) that is not well conserved among nuclear receptors; an evolutionarily conserved DNA binding domain (DBD); a hinge region (region D); and a C-terminal ligand binding domain (region E or LBD). Region E integrates multiple functions. In addition to determining the ligand-binding properties of a particular receptor, this region also specifies its dimerization properties (homo- versus heterodimerization) and contains a ligand-dependent transactivation function (activation function 2 (AF-2)). The crystal structures of E regions of several hormone receptors reveal a similar fold in LBDs, which may be a common property of all NRs (2-4). This newly determined fold, termed the "antiparallel -helical sandwich," consists of 12 -helices (H1-H12) packed in three layers, ultimately generating an internal hydrophobic ligand-binding cavity. The striking difference between the ligand-bound and unbound LBD is the position of H12 (AF-2 helix). In the absence of the ligand, the helix projects away from LBD but tightly folds toward the LBD when ligand is bound. It is proposed that this conformational change may yield an interacting surface for NR coactivators (5).

In addition to a weak dimerization surface defined in DBD, a 40-amino acid region has been mapped as a dimerization interface (identity box (I-box)) within the carboxyl portion of LBDs of RAR, TR, COUP, and RXR (6, 7). Coincidentally, this region matches almost perfectly with a helical segment formed by H9 and H10, whereas a dimerization interface is localized mainly in H10 of the LBD crystal structures (2, 4). These experiments have revived interest in the ninth heptad, because it is one of the nine heptad repeats that were predicted to organize a leucine zipper-like structure mediating dimerization in RAR, VDR, and T3R (8-10), and heptad 9 is contained within H10.

For a large number of orphan nuclear receptors, high affinity endogenous ligands have not been found, although many of them exert transcriptional activities. Steroidogenic factor 1 (SF-1) belongs to this group (11-13). SF-1 is a homologue of Drosophila Ftz-F1 (14, 15). Playing a key role in the development and differentiation of the adrenal gland, gonads, ventromedial hypothalamus, and pituitary gonadotrope (16-20), mouse SF-1 (mSF-1) is a transcription factor regulating a variety of genes (Ref. 21; summarized in Ref. 22). Unlike many hormone nuclear receptors, mSF-1 binds to its DNA element as a monomer (23, 24). In addition, transient expression study revealed that truncation of the entire LBD of mSF-1 resulted in a constitutive activator of the Mülerian inhibiting substance gene promoter (17). Thus, the mechanism of mSF-1 in transcription could be different from that of other dimeric nuclear receptors.

Our laboratory previously identified the zebrafish Ftz-F1 homologue, zFF1A (zebrafish Ff1a or Nr5A2 (25)) and its naturally truncated variant, zFF1B. The lack of an I-box and AF-2 in zFF1B presumably abolishes its activity and its failure to act in synergy with ligand-bound ER on the salmon gonadotropin II subunit gene (sGTHII) promoter (6). In the present study, we show that deletion of the I-box or mutating it at heptad 9 drastically reduces the transcriptional activity of zFF1A, but such a reduction is less pronounced in the zFF1/ER synergy. Furthermore, wild-type zFF1A and its mutants all retained interaction with ER-LBD. To delineate the transactivation functions of zFF1A, we have defined a new activation function in the D domain (AF-3) and a negative region between AF-3 and the I-box. Our results suggest an unusual role of the "dimerization domain" in Ftz-F1 homologues and suggest that zFF1A transactivation is due to a coordination of multiple subdomains in the D and E regions.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Primers-- All oligonucleotide primers were synthesized in Biotechnology Service Center, the Hospital for Sick Children (Toronto, Canada). Primers designed for the zFF1A and B ORF amplification were as follows: 5'(A/B)Bm, AGAGGATCCATGCTGCCTAAAGTCG; 3'(A)Eo, GCGAATTCTCAGGCACGTTTGG; 3'(B)Eo, CCCAATTCTCTCTACTCACCCAAGC; and 3'(A)Hn, GGCAAGCTTTCAGGCACGTTTGGCG (where either BamHI (Bm), EcoRI (Eo), or HindIII (Hn) restriction site is underlined). Primers used for C-terminal deletions of zFF1A protein were as follows: 1-501, AAGAGAATTCTCAGGGCACGTCTCCGTTCAGGTGT; 1-155, AGTGAATTCGCCCGGATCAAGGCTTTCT; 1-321, CATTAGGCCGAAAGTGTTGAG. The primer used to obtain the N- terminal deletions (43-516, 43-155, 43-321, and 43-434) of zFF1 was (1-43), GACGGATCCATGGATGAGATGTGTCC. The I-box internal deletion primers were AatII upstream (TCTTGACGTCCAAGCTGAACAGCA and AatII downstream (GCCGACGTCTACCTGTACTACAAAC) (where the AatII restriction sites are underlined) and T3 and T7 primers of pBluescript (Stratagene, La Jolla, CA). The point mutation primers designed to test leucine residues in heptads 8 and 9 of zFF1A were as follows: mutIB-1, AGACAAGTTCGGCCAGGCGGCGCTGCGGCTGCCGGAG; mutIB-3, CCAGCTGCTGCTGCGGGCGGCGGAGATCCGCGCCAT; and mutIB-4, AGCAGGTGAACGCGGCAGCGGCGGACTACGTGATGTGC (where the underlines indicate the mutated nucleotides). For internal deletions mutants of the putative inhibitory region, the following primers were employed: (1)Hn, GCAAAGCTTATGCTGCCTAAAG; TC; (321)Bm, AGCGGATCCGCACATTAGGCC; (412)Bm, TGCGGATCCGCTCAGCAGCTC. Primers used to PCR-amplify the DNA fragments encoding the D, E, I-box, and AF-2 regions of zFF1A and their combinations were as follows: (155)Eo, AAGGAATTCGCCATGACTCAAGTC; (267)Bm, TGCGGATCCTTCTACGTAAGGGTA; (278)Eo, TCGGAATTCTCCTTCCCTCACTTA; (322)Eo, CTAGAATTCAAGATGGCTGACCAG; (436)Bm, TCAGGATCCACGTGAAGAACCTGG; (498)Eo, AAAGAATTCAACGGAGACGTGCCC; (516)Xa, GGCTCTAGATCAGGCACGTTTGGC (where the BamHI (Bm), EcoRI (Eo), or XbaI (Xa) site is underlined). Primers used for sequencing confirmation of the insert/vector junctions of GAL4/zFF1A fusion constructs were based on pM plasmid (CLONTECH).

Constructs and PCR-- The zFF1A and B cDNAs, from the original library screening (2), served as the templates for amplification of their putative open reading frames by PCR, and the primer 5'(A/B)Bm in combination with either 3'(A)Eo or 3'(B)Eo primer was employed. The resulted ORF DNA fragments were subcloned into the eukaryotic expression vector pCDNA3 (Invitrogen, San Diego, CA) at its BamHI and EcoRI sites and named as A-ORF and B-ORF. The A-ORF plasmid then served as template for construction of deletions by PCR. For C-terminal series deletion, the upper strand primer was 5'(A/B)Bm, and the lower strand primer included primers 1-516, 1-321, and 1-155. As a consequence of PCR and subcloning, deletions 1-516, 1-321, and 1-155 were obtained. The construction of the deletion 1-267 included two steps; the zFF1A ORF was first amplified by primers 5'(A/B)Bm and 3'(A)Hn and ligated into pCDNA3 at the BamHI and EcoRV sites. Next, the resultant plasmid was treated by HindIII restriction to eliminate a region between 266 and 516. After ligation, deletion 1-267 was obtained. For N-terminal deletion, primer (1-43) was used, in combination with either primer 3'(A)Eo, 1-321, 1-155, or 3'(B)Eo. The PCR-amplified DNA fragments were then restricted by BamHI and EcoRI and cloned back into pCDNA3. Deletion 1-434 was equivalent to zFF1BORF. The internal deletion A-I-box was produced by three steps. First, the AatII upstream and downstream primers were combined with either T3 or T7 primers of pBluescript to PCR-amplify two fragments from full-length zFF1A cDNA. An AatII restriction enzyme site was included at 5'-ends of both primers. Second, the DNA fragments with correct sizes were treated by AatII restriction and ligated using T4 ligase (Life Technologies, Inc.). Finally, one-tenth of the ligation mixture was taken to perform PCR with primers 5'(A/B)Bm and 3'(A)Eo and the expected fragment A-I-box was 156 base pairs shorter than wild type or its predicted ORF, which should contain 462 amino acids without the entire I-box. Point mutations were introduced into A-ORF by a PCR-based mutagenesis method described elsewhere (26). Principally, a mutagenic primer (mutIB-1, mutIB-3, or mutIB-4) was designed in the region where mutation was introduced, and following PCR with primer 3'(A)Eo, a large DNA fragment (~130-200 base pairs) was obtained. Using the "large fragment" primer, which was treated by DpnI digestion to remove the possible parental plasmid contamination, and primer 5'(A/B)Bm, the full-length ORF fragment with the expected mutation was obtained by PCR. The mutated ORF DNA fragment was then restricted by BamHI and EcoRI and ligated into pCDNA3. The mutated region was confirmed by DNA sequencing and restriction mapping. No unexpected mutations were found. About 20 ng of A-ORF template and 2 units of Pfu polymerase (Stratagene, La Jolla, CA) were used in each PCR, and the amplification procedure generally consisted of 3 min of 95 °C followed by 20 cycles at 93 °C (1 min), 60 °C (1 min), and 75 °C (2 min). In certain cases, such as the introduction of the point mutations and the generation of the internal deletion, annealing and extension temperatures were adjusted according to the T_m of the primers used. Using PCR amplification of zFF1 AORF, the internal deletion construct zFF1 IR was created by fusion of (1)Hn/(321)Bm and (436)Bm/(516)Xa.

The pG5CAT reporter and pM fusion vector were from a mammalian two-hybrid assay kit (CLONTECH). 155-516 represents an DNA insert encoding the entire D and E regions of zFF1A, which was PCR-amplified from zFF1A cDNA by primers (155)Eo/(516)Xa. To obtain the individual D and E encoding fragments, primers (155)Eo/(267)Bm and (278)Eo/(516)Xa were employed in the PCRs, respectively. Primers (436)Bm/(516)Xa and (498)Eo/(516)Xa amplified I-box plus AF2 and AF-2 fragments, respectively, and ligation of the I-box plus AF2 and D fragments resulted in the internal deletion of the N-terminal portion of the E domain. To include Regions II and III with I-box plus AF2, primers (322)Eo/(516)Xa were used instead of (436)Bm/(516)Xa. The internal deletion fragment DE IR was made by ligating fragments amplified by primers (155)Eo/(321)Bm and (436)Bm/(516)Xa. For all subcloning, the PCR products were digested with the respective enzymes and ligated in frame at the C terminus of the Gal4 DBD. All of the PCR-derived clones were confirmed by sequencing on both strands.

LexA fusion vectors to express ER LBD and ER, along with GST fusion vector expressing ER LBD, were previously described (27). A PCR-amplified fragment of mSF-1 (28) was subcloned into EcoRI-XhoI restriction sites of the B42 fusion vector pJG4-5 (29).

Yeast Two-hybrid Test-- For the yeast two-hybrid tests, plasmids encoding LexA fusions and B42 fusions were cotransformed into Saccharomyces cerevisiae EGY48 strain containing the LacZ reporter plasmid, SH/18-34 (29). Plate and liquid assays of -galactosidase expression were carried out as described (29). Similar results were obtained in more than two similar experiments.

GST Pull-down Assay-- The GST-ER-LBD fusion or GST alone was expressed in Escherichia coli, bound to glutathione-Sepharose beads (Amersham Pharmacia Biotech) and incubated with labeled zFF1 proteins expressed by in vitro translation using the TNT-coupled transcription-translation system, with conditions described by the manufacturer (Promega, Madison, WI). Specifically bound proteins were eluted from the beads with 40 mM reduced glutathione in 50 mM Tris (pH 8.0) and analyzed by SDS-PAGE and autoradiography as described (29).

Transfection and CAT Assay-- Plasmids such as the reporter CAT constructs, mSF-1, and rainbow trout ER expression constructs were reported previously (2, 22, 30, 31). The SRC-1a and TRAM-1 expression constructs were gifts from Dr. M-J. Tsai (Baylor College of Medicine, Houston) and Dr. A. Takeshita (Brigham and Women's Hospital, Boston), respectively. Cell culture, transfection, hormone treatment, and CAT assay were performed essentially as described previously (22). In general, 1 µg of transcription factor expression plasmid, 5-10 µg of CAT reporter plasmid, and 1 µg of pCMV-gal plasmid (as internal control) were included in each transfection, or the manufacturer's instructions (CLONTECH) were followed. Coactivator expression plasmid or an equal amount of pBluescript was added to test the effect of the coactivator. All cells were grown in 6-cm dishes.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The I-box May Regulate Interaction of zFF1A and ER by Hydrophobic Contact-- We previously showed that both mSF-1 and zFF1A were able to synergize with ligand-bound ER to up-regulate sGTHII promoter activity (6, 21). To verify the nature of the synergy, a yeast two-hybrid test showed that mSF-1 interacted with both types of estrogen receptor, ER and ER (Table I). The interactions were estradiol (E₂)-independent, suggesting that repositioning of ER AF-2 in the presence of ligand is not essential for contact.

View this table: [in this window] [in a new window]   Table I Interactions of SF-1 with ER and ER in yeast The indicated B42 and LexA plasmids were transformed into a yeast strain containing an appropriate lacZ reporter gene. At least six separate transformants from each transformation were transferred to indicator plates containing 5-bromo-4-chloro-5-indolyl -D-galactopyranoside (X-gal), and reproducible results were obtained using colonies from a separate transformation. ++, strongly blue colonies after 2 days of incubation and strong interaction; +, light blue colonies after 2 days of incubation and intermediate to weak interaction; , white colonies and no interaction.

Because zFF1B lacked a synergistic interaction with ER, we postulated that the interaction between zFF1A and ER might be at least partially mediated by the I-box (6). We compared all vertebrate Ftz-F1 homologues with other nuclear receptors that can form dimers, in the area of the last several helices near the C terminus of LBDs (H9-H10, and H12). Among the known Ftz-F1s, in addition to the completely conserved AF-2 core, the regions corresponding to the I-box are also conserved (Fig. 1). The overall conservation of the I-box regions between Ftz-F1 homologues and other nuclear receptors is moderate but is higher in the H10 region, especially in the hydrophobic region of heptad 9. Comparison of zFF1A heptad 9 with that of other NRs including ER revealed a high degree of conservation among the 9 amino acids (see boxed region of heptad 9 in Fig. 1).

View larger version (55K): [in this window] [in a new window]   Fig. 1.   Heptad 9 in the I-box of zFF1A. Comparison of Ftz-F1 members and other nuclear receptors in H9, H10, and H12 regions are shown. In addition to the complete conserved AF-2 core region in all vertebrate Ftz-F1 homologous members, the heptad 9 region also shows a considerably high similarity. This high identity is best exampled when zFF1A, xFF1rA, RXR, dUSP, and COUP-TF1 are compared. The alignment of NRs other than Ftz-F1s in the region was derived from Wurtz et al. (4).

Both AF-2 and the I-box Determine the Major Activation Function of zFF1A-- From cotransfection studies, it appeared that most of the C-terminal 82 amino acids (positions 434-516) in zFF1A, including the AF-2 helix, were critical for the zFF1A and ER synergy. In the absence of AF-2 (AF-2), the synergistic effect of zFF1A and ER on sGTHII promoter was no longer significant (Fig. 2B). In the absence of both AF-2 and the I-box, the synergistic effect mediated by ER and zFF1 was lost. We thus constructed an internal deletion, A(I-box), in which only the I-box was removed (Fig. 2A). Cotransfection in COS-1 cells revealed that in the presence of AF-2, the synergy was barely observed when A(I-box) was tested (similar to B-ORF, Fig. 2B), suggesting that the I-box may mediate interaction with ER. In contrast, in the absence of ER, deletion of either AF-2 or the I-box also abolished the activation function of zFF1A on the sGTHII promoter (Fig. 2C), further highlighting the critical role of the I-box for AF-2 activity.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   The I-box, especially heptad 9, is crucial for the transcription activity of zFF1A. A, schematic structures of wild-type and mutated zFF1A used in transfection study and GST pull-down assay. B, the role of AF-2, I-box, and heptad 9 in the zFF1A and ER/E2 synergistic effect on target promoter. A sGTHII-1260/CAT reporter was used in cotransfection (2, 22), and the results shown are from three independent tests in COS-1 cells (in duplicate). Basal is defined as the reporter activity in the absence of any additional transcription factor. C, transcriptional activities of wild-type and mutated proteins in the absence of ER. Basal is defined as the promoter activity in the presence of ER/E2 only.

Heptad 9 (Amino Acids 474-482) in H10 Is Required for zFF1A Transcriptional Activity-- Point mutations were introduced within the conservation shown in Fig. 1 to assess the role of specific amino acids in the I-box. Two mutants were constructed to analyze the function of heptad 9. In L474A,L475A, leucines 474 and 475 were substituted by an alanine, and in L478A,P479A, leucine 478 and proline 479 were substituted with alanine. Mutant L456A,L457A, in which two leucine residues of heptad 8 were converted into alanine, served as a control (Fig. 2A). Only mutant L474A,L475A, in the presence of ER/E₂, was unable to stimulate sGTHII promoter activity further, whereas the other two mutants exhibited function similar to that of the wild-type protein, i.e. they cooperatively enhanced the activity of the sGTHII gene promoter with ER (Fig. 2B).

Mutant L474A/L475A showed transactivation activities far below that of the wild-type protein in the absence of ER/E₂ (Fig. 2C). Thus, deletion of the AF-2 or I-box or change of the first double leucine residues in heptad 9 all resulted in the complete loss of transactivation function of zFF1A, confirming that the AF-2 and I-box helices are functionally related.

The I-box and AF-2 Helices Do Not Mediate Direct zFF1A/ER Interaction-- To test whether the interaction between ER and zFF1A or zFF1A mutants could be attributed to the heptad 9 or the I-box, glutathione S-transferase-ER-LBD was incubated with zFF1A or its mutants were labeled with [³⁵S]methionine by in vitro translation. All zFF1 proteins (A-ORF, A(AF-2), A(I-box), L474A,L475A, L456A,L457A, and B-ORF) showed physical contact with ER that was independent of E₂ (Fig. 3), similar to that of the mSF-1/ER interaction (Table I). These data demonstrated that neither the I-box nor heptad 9 mediates the zFF1A and ER contact and suggested that the major interacting surfaces of zFF1A are probably outside of the I-box and AF-2.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   All zFF1A proteins are able to physically contact with ER-LBD in vitro. The 35S-labeled zFF1 proteins expressed by in vitro translation were incubated with GST-ER-LBD bound to glutathione-Sepharose-4B beads. Specific binding proteins were eluted and analyzed by SDS-PAGE and autoradiography. 35S-Labeled luciferase and GST alone served as negative control. The presence or absence of E2 in the reactions essentially gave rise to the same results.

Other Regions of zFF1A Are Required for Transcriptional Activity-- To delineate regions corresponding to the full transactivation of zFF1A, a series of C-terminal deletion constructs were made (Fig. 4). The removal of the AF-2 region resulted in a significant loss of the transcriptional activity of zFF1A (as seen in deletion 1-501). Further truncation of 67 amino acids from the C terminus completely abolished the activity of zFF1A (deletion 1-434), whereas deletion to amino acid 155 (deletion construct 1-155) resulted in transactivation similar to that of deletion 1-501. This result suggested that the AF-2 is critical to the activation function of zFF1A, whereas the contribution of the A/B region is minor. Such a notion was supported by the N-terminal truncation of the A/B region (deletion construct 43-516), because the deletion did not change the activity of zFF1A. Only when the AF-2 motif and A/B region were deleted simultaneously (deletion construct 43-434) was the activity of zFF1A drastically reduced (Fig. 4A).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Delineation of the activation function regions in zFF1A. A, a series of deletion constructs are shown on the left (scheme is not to scale), and the numbers indicate the truncation sites. All constructs were tested through a cotransfection study in HeLa cells. The reporter construct was the sGTHII gene 39 basal promoter linked to CAT, with a consensus GSE sequence 5' upstream (consGSE-TATA-CAT) (2). Basal value represents the activity of the consGSE-TATA promoter without the addition of any transcription factor. B, the same zFF1 constructs were tested for their synergistic action with ER/E2 on a sGTHII kilobase pair 0.4 promoter (22). Cells cotransfected with ER were treated by E2 (1 µM) after transfection. Basal activity refers to that of the promoter in the presence of ER/E2 only. All histograms represent the mean ± S.D. of at least two independent experiments (in duplicate or triplicate).

An apparent increase in activity of zFF1 was obtained by C-terminal deletion to amino acid 267 (deletion 1-267), whereas further deletion abolished the regained activity (deletion construct 1-155). Therefore, the activity of deletion 1-267 was due to a region between 155 and 267, which we named activation function 3, or AF-3. AF-3 could act independently of the A/B region as demonstrated by deletion 43-321 (Fig. 4A).

To analyze whether the AFs and A/B region are involved in the synergistic effect mediated by zFF1A with ER, all of the deletion constructs were tested in HeLa cells in the presence of ER/E₂ (Fig. 4B). In addition to the full-length zFF1A (residues 1-516), deletions 1-501 (with A/B region and AF-3), 1-267 (A/B region and AF-3), 43-321 (AF-3), and 43-516 (AF-2 and AF-3) still exerted variable synergy with ER, suggesting that both AF-2 and AF-3 contributed to the synergy. The A/B region alone (deletion construct 1-155) did not show any synergy, but its removal in deletions 43-516 and 43-321 decreased the synergy. This decrease could be due to the weak transcriptional activity of the A/B region (AF-1), whereas deletion 1-155 slightly enhanced the promoter activity above the activity with ER alone (basal). The lack of any AFs might explain why deletion of 43-155 was silent.

The region between 267 and 434, which encompasses regions II and III, could serve as an inhibitor of AF-3 (deletion 43-434, Fig. 4A). The loss of synergy by deletions 1-434 and 43-434 might also reflect the inhibitory effect of region 267-434 (Fig. 4B). However, without the inhibitory region (IR), A(IR) did not transactivate as efficiently as AORF. On the other hand, the synergistic effect mediated by AFs of zFF1 and ER was not affected by the removal of the IR (Fig. 5). The overall organization of the regions related to zFF1A transcriptional activity is shown in Fig. 6A.

View larger version (38K): [in this window] [in a new window]   Fig. 5.   The IR affects transactivation. The left panel shows that the IR deletion (A(IR)) results in the decreased activity of consGSE-TATA promoter (compared with AORF). The same deletion did not affect its synergistic action with ER on a sGTHII kilobase pair 0.4 promoter (right panel). All histograms represent the mean ± S.D. of at least three independent experiments (in triplicate). Other details are the same as in Fig. 4.

View larger version (31K): [in this window] [in a new window]   Fig. 6.   Multiple regions are required for the transactivation of zFF1A. A, a schematic diagram of regions related to the transactivation of zFF1A is presented (from the results shown in Figs. 4 and 5), and the D/E region is further enlarged (lower panel). Numbers represent the sites where the subdomains of the D/E region were originally defined in Fig. 4B. Individual AFs and their combinations were fused to GAL4-DBD and tested by cotransfection with pG5CAT. The results are from at least three independent experiments (in duplicate each time), and the histograms represent mean ± S.D. The CAT activity of reporters obtained by transfecting GAL4/155-516 is considered 1, and the other CAT activity values are shown as percentages relative to 1.

The Transcriptional Activity of zFF1 Depends on Several Activating Functions-- We further fused individual AFs and their flanking regions (Fig. 6A) to GAL4-DBD. The fusion constructs were tested in HeLa cells, with the E1b-CAT reporter driven by five GAL4 binding sites. As summarized in Fig. 6B, consistent with our earlier observations (above), the I-box was essential for AF-2 activity (I-box plus AF-2; GAL4/436-516 versus AF-2 alone; GAL4/498-516), and the AF-3 (GAL4/155-267) was an independent activation domain. The transactivation of either I-box plus AF-2 or AF-3 was not comparable with that of the entire D/E region (GAL4/155-516). In addition, the entire E region (GAL4/278-516) also showed limited activity, comparable with that of the I-box plus AF-2 alone. These data suggested a relatively weak activity of each AF and a synergistic requirement of all AFs for the full activation of zFF1A. However, internal deletion of the IR in GAL4/155-267;436-516 (AF-3, I-box, and AF2 only) resulted in less transactivation than that of GAL4/155-516. The lack of the intact IR in GAL4/322-516 appeared to eliminate the activity of the I-box plus AF-2.

The I-box Mediates SRC-1 Recruitment to zFF1A-- The role of the steroid receptor coactivator 1 (SRC-1) in potentiating the transactivation of zFF1 was examined. In the presence of E₂, SRC-1 could potentiate ER activity on the sGTHII gene proximal ERE/TATA promoter by more than 3-fold, an effect that was described previously for an ERE-containing promoter (32). SRC-1 could potentiate both AORF and mSF-1 function on the consGSE-TATA promoter by 2-fold, whereas no effect was evident on BORF (Fig. 7). As well, SRC-1 failed to enhance the activity of A(I-box) significantly but had an effect on A(AF2) (data not shown). Therefore, without the I-box, zFF1 might not respond efficiently to the coactivator.

View larger version (42K): [in this window] [in a new window]   Fig. 7.   SRC-1 selectively activates zFF1A. SRC-1 could significantly potentiate the function of zFF1A, mSF-1, and rainbow trout ER but showed limited effect on zFF1B. All results represent three independent experiments in HeLa cells (in duplicate), and 5 µg of SRC-1 expression vector was used in each transfection.

To determine the role of SRC-1 in the synergistic interaction of zFF1A/ER, AORF was cotransfected with both ER/E₂ and SRC-1. There was a 2-fold increase of the synergistic effect of AORF/ER (on sGTHII promoter) by SRC-1 (Fig. 8). A greater potentiation of SRC-1 on the activity of A(I-box) (3.5-fold) and A(AF2) (4.7-fold) was observed, but no significant enhancement was evident on BORF. The presence of the I-box in A(AF2) apparently led to a much higher promoter activity (with ER and ER/SRC-1) than that induced by A(I-box), indicating that although both the I-box and AF-2 exerted a relatively weak interaction with SRC-1 (compared with ER), the I-box might have a higher affinity for the coactivator. HeLa cells express endogenous SRC-1 (33). In our test, cotransfecting large amounts of SRC-1 expression construct would result in a much higher level of SRC-1 in cells. These data therefore suggested that overexpression of SRC-1 in cells and its dominant recruitment by ER to the target promoter might contribute to the more pronounced potentiation of A(AF2) and A(I-box) by the coactivator.

View larger version (36K): [in this window] [in a new window]   Fig. 8.   The I-box mediates the function of SRC-1. Through cotransfection study with ER/E2 and/or SRC-1 in HeLa cells, the role of the I-box in mediating SRC-1 function has been evident. The -fold induction is relative to the promoter activity in the absence of any transcription factors and coactivator. The reporter construct is a sGTHII kilobase pair 0.4 promoter (22). All histograms represent the mean ± S.D. of at least three independent experiments (in duplicate). The numbers above the bars are the -fold potentiation of SRC-1. SRC-1 alone showed no significant effect on the promoter.

In the presence of ER/SRC-1, A(IR) also exerted greater potentiation (4.3-fold) and led to a promoter activity comparable with that of AORF (69-fold versus 73-fold). However, without the exogenous SRC-1 or lacking a strong recruitment of SRC-1 by ER to the sGTHII promoter, A(IR) was not as efficient as AORF in transactivation (Fig. 5). Thus, it appears that the IR is unlikely to make direct protein-protein contacts with SRC-1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

To test the function of the zFF1A I-box, we have determined that heptad 9 of zFF1A is required for AF-2 activity, although earlier mutation studies indicated that several hydrophobic residues in heptad 9 of TR or RAR could determine their interaction with RXR (34-36). We have further shown that the I-box and AF-2 of zFF1 are sufficient to recruit SRC-1. We have also identified a novel activation function, AF-3, that spans most of the D region of zFF1A (residues 155-266), and we have delineated a region (residues 267-434) that may specifically inhibit AF-3 activity, although the nature of inhibition is still unclear. These results represent the first demonstration that both the heptad 9 and hinge region, in addition to AF-2, are involved in the transcriptional activity of a nuclear receptor with a monomeric DNA binding feature. Our data stresses the notion that interdomain communication (37), mediated by three segments (AF-3, I-box, and AF-2) in the D/E region of zFF1A, confers the full transcriptional activity of a nuclear receptor.

Our study of the I-box was prompted by the determination of the crystal structure of nuclear receptor LBDs and the evidence that mSF-1 interacts with ERs (Table I). Consistent with a dimerization interface found in H10 of the RXR LBD (2), the I-box has been mapped to the H9-H10 region in LBDs of RXR, RAR, TR, and COUP and mediates heterodimerization (7). In addition, the function of the I-box in the homodimerization of HNF4 was also determined, although there is only moderate conservation of the I-box between RXR and HNF4 (9). Furthermore, ER has been proposed as a common interaction partner for TR, RAR, RXR, and HNF4 via their LBDs (27). The loss of function of zFF1B in synergy, due to deletion of both AF-2 and the I-box, thus initially suggested that the I-box might contribute to the interaction of zFF1A and ER either directly or indirectly through other intermediary proteins. It now appears that rather than providing an ER-interacting interface, the I-box, especially heptad 9 of zFF1A, is directly related to AF-2 activity. Our GST pull-down assay further rules out the possibility that the physical contact between zFF1A and ER depends directly on the AF-2, I-box, or heptad 9. Further experimentation is needed to determine which domain of zFF1 mediates protein-protein interactions with ER.

Both SRC-1 and TRAM-1 (thyroid hormone receptor activator molecule (25, 33)) can potentiate zFF1A activity, but they show no effect on zFF1B (Fig. 4 and data not shown). Furthermore, only A(I-box) and mutation L474A,L475A failed to respond to the co-expression of SRC-1 (data not shown), arguing an important role of the I-box in mediating the activation of coactivators. From the cotransfection of SRC-1, zFF1 mutants, and ER, the role of the I box in coactivator recruitment becomes more evident. Although the ER should play a significant role in SRC-1 recruitment (32), the removal of the I-box in A(I-box) and BORF appears to disrupt the coactivator/zFF1 interaction. Consequently, the combinatory effect of transcription factors and SRC-1 on the sGTHII promoter is much higher only when the I-box is present (Fig. 8). However, in the absence of exogenous SRC-1, the sGTHII promoter activity is also higher when zFF1s with the I-box are employed (A(I-box) versus A(AF2), AORF, or A(IR) in Fig. 8), suggesting the involvement of the endogenous SRC-1 or other coactivator(s). The SRC family represents a class of general coactivators that are able to modify chromatin structure, while the docking of specific coactivators to DNA-binding transcription factors recruits the general coactivators (38). Whether HeLa cells express Ftz-F1-specific coactivator(s) is currently unknown, but the endogenous SRC-1 may substitute for such factor(s) that are likely to be missing. The overproduction of SRC-1 in cells would exert a broad effect shown for many NRs, on zFF1s via ER, since ER and zFF1 are bound to each other (our GST pull-down result).

In the elucidation of the nuclear receptor-coactivator complexes assembled in response to hormonal signals, several leucine-rich motifs (LXXLL) were found in the nuclear receptor interaction domains of SRC family members (39, 40). In SRC-1, three LXXLL motifs have been shown to mediate direct contact with ligand-bound steroid receptors. Because a leucine-rich motif in cyclin D1 is able to recruit the coactivator to ER in the absence of E₂ (41), it is possible that SRC-1 interacts with any such motifs on a given receptor. Since the heptad 9 is leucine-rich, highly hydrophobic, and conserved among Ftz-F1s (Fig. 1), it is not surprising that lack of this segment compromises the transcriptional activity of zFF1 even in the presence of exogenous SRC-1.

A proximal activation function domain resides between residues 181 and 310 of xFF1rA (42), and a domain mapped upstream of AF-2 (residues 187-245) in mSF-1 was found to be essential for their transcriptional activity (43, 44). These domains span parts of the D and E regions. According to a sequence alignment of the LBDs in all nuclear receptors, both domains overlap with the AF-2a domain of ER and the 2 region of GR in the LBDs (4, 45, 46). The proximal activation domains of xFF1rA and mSF-1 are capable of transactivation when fused to a heterologous DBD. The AF-2 of mSF-1, on the other hand, lacks any independent activity. It was concluded that AF-2 is necessary, but insufficient, for the transcriptional activity of the vertebrate Ftz-F1 homologues (42, 43).

Our current study indicates that when linked to its own or a heterologous DBD, the AF-3 (encompassing the entire D region), without any overlap with the N terminus of the E region, can serve as an activation domain. As expected, when tethered to GAL4-DBD, AF-2 alone is not active at all. Moreover, the I-box and AF-2 together can confer the major transactivation function of the E region. Our data further indicate that without the I-box (A(I-box)), or with an altered heptad 9 (L474A,L475A), the AF-2 activity of zFF1A is barely detectable. The weaker activity of each AF and the greater transcriptional activity of the D/E region have led to a speculation that the full transcriptional activity of zFF1A requires all three segments (Figs. 5 and 6). By testing the synergistic interaction of zFF1A and ER using the C- and N-terminal deletions of zFF1A, both AF-2 and AF-3 of zFF1A were found to make significant contributions to transactivation. In fact, it appears that the D region not only harbors an activation domain but is also necessary for the interaction of both nuclear receptors (zFF1A and ER).

It is intriguing to note that deletion of the IR region significantly reduces zFF1 transactivation but does not appear to function in coactivator recruitment. Therefore, the inability of zFF1B to respond to SRC-1 is not due to the inhibitory domain per se but rather the lack of both the I box and AF2. Alternatively, the IR segment could simply act as a linker to ensure a proper comformational change of the D/E region and communication between the AF-3 and I-box plus AF-2, because partially deleting the IR results in a more severe reduction.² However, we cannot rule out that the corepressor recruitment and/or post-translational modification can be mediated by the IR. In fact, the modifications play an important role in mSF-1 transcriptional activity.

No apparent ligand is needed for mSF-1 activity in many cellular settings (i.e. mSF-1 is functional in various cell lines). Since evidence that oxysterols can serve as a bona fide ligand remains controversial (12, 13), it is thought that mSF-1 may be regulated by alternate mechanisms involving coactivators, corepressors, and signal transduction pathways.

Such a notion recently received support, because a constitutive phosphorylation site (Ser²⁰³) in region 187-245, which can be modified by the mitogen-activated protein kinase signaling pathway, has been demonstrated to maximize the mSF-1-mediated transcription (47). Like phosphoserines often found in the A/B regions of many hormone NRs, phosphorylation of the hinge region by mitogen-activated protein kinase results in ligand-independent activation and recruitment of NR cofactors (47, 48). The phosphorylation of Ser²⁰³ in the proximal AF of mSF-1 increases the binding of corepressor SMRT (silencing mediator of retinoic acid and thyroid hormone receptor) but shows less effect on its coactivator association in vitro (47).

Interestingly, a serine/threonine mitogen-activated protein kinase phosphorylation site PFVTSP (resembles the PX_n(S/T)P consensus) is also found in region 155-267 of zFF1, but the location of the mitogen-activated protein kinase Ser/Pro consensus is quite different from that in mSF-1. Therefore, we suggest that the recruitment of NR corepressors by AF-3 is likely and should be important for the activity of zFF1B. Because NR corepressors can also directly interact with the E regions of many NRs (49), the possibility that the inhibitory region (residues 267-434) may be responsible for the corepressor recruitment to zFF1 is not excluded.

The proximal AF and LBD of mSF-1 are required for the interaction with either SMRT or coactivators (43, 47). Therefore, mSF-1 transactivation may be determined by both coactivators and corepressors in vivo. Whether the same mechanism applies to zFF1 requires further investigation. Nevertheless, the contribution of multiple subdomains in the D/E region, especially the I-box, to the zFF1A transactivation may represent a general mechanism of Ftz-F1 function.

	ACKNOWLEDGEMENTS

We especially thank Dr. Bon-chu Chung (Institute of Molecular Biology, Academia Sinica, Taiwan) for sharing unpublished data. We are grateful to Drs. H. P. Elsholtz, C-C. Hui, M-J. Tsai, and B. Schimmer for helpful suggestions and technical advice. We also thank Drs. P. Melamed and M. Westerfield for critical reading of the manuscript. We appreciate the technical assistance of L. Liao and J. Ando and thank L. Mark for helping in the preparation of the manuscript.

	FOOTNOTES

^* This work was supported by Medical Research Council of Canada Grant MT-12900 (to C. L. H.) and by a grant from the National Creative Research Initiative program of the Korean Ministry of Science and Technology (to J. W. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^c Recipient of the Restracom Trainee Fellowship. Present address: Institute of Neuroscience, University of Oregon, Eugene, OR 97403-1254.

^d Recipient of a fellowship from the Fondation J. Armand Bombardier.

^f Visiting scientist, partially supported by the Hospital for Sick Children. Present address: Biologie Cellulaire & Reproduction, University of Rennes, Rennes Cedex, 35042 France.

^g Present address: Graduate School of Science, Hokkaido University, Sapporo, Hokkaido 060-08100, Japan.

^j To whom correspondence should be addressed: Dept. of Laboratory Medicine and Pathobiology, University of Toronto 100 College St., Rm.

Published, JBC Papers in Press, March 20, 2000, DOI 10.1074/jbc.M000121200

² M. Chandy and C. L. Hew, unpublished data.

351, Toronto, Ontario M5G 1L5, Canada. Tel.: 416-978-6505; Fax: 416-978-8802; E-mail: choy.hew@utoronto.ca.

	ABBREVIATIONS

The abbreviations used are: NR, nuclear receptor; AF, activation function; DBD, DNA binding domain; LBD, ligand binding domain; SF-1, steroidogenic factor 1; mSF-1, mouse SF-1; ER, estrogen receptor; ORF, open reading frame; GST, glutathione S-transferase; CAT, chloramphenical acetyltransferase; E₂, estradiol; IR, inhibitory region.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. Qin, D.-m. Gao, Q.-F. Jiang, Q. Zhou, Y.-Y. Kong, Y. Wang, and Y.-H. Xie Prospero-Related Homeobox (Prox1) Is a Corepressor of Human Liver Receptor Homolog-1 and Suppresses the Transcription of the Cholesterol 7-{alpha}-Hydroxylase Gene Mol. Endocrinol., October 1, 2004; 18(10): 2424 - 2439. [Abstract] [Full Text] [PDF]

				P.-L. Xu, Y.-Q. Liu, S.-F. Shan, Y.-Y. Kong, Q. Zhou, M. Li, J.-P. Ding, Y.-H. Xie, and Y. Wang Molecular Mechanism for the Potentiation of the Transcriptional Activity of Human Liver Receptor Homolog 1 by Steroid Receptor Coactivator-1 Mol. Endocrinol., August 1, 2004; 18(8): 1887 - 1905. [Abstract] [Full Text] [PDF]

				Y.-W. Liu, W. Gao, H.-L. Teh, J.-H. Tan, and W.-K. Chan Prox1 Is a Novel Coregulator of Ff1b and Is Involved in the Embryonic Development of the Zebra Fish Interrenal Primordium Mol. Cell. Biol., October 15, 2003; 23(20): 7243 - 7255. [Abstract] [Full Text] [PDF]

				Y. Yoshiura, B. Senthilkumaran, M. Watanabe, Y. Oba, T. Kobayashi, and Y. Nagahama Synergistic Expression of Ad4BP/SF-1 and Cytochrome P-450 Aromatase (Ovarian Type) in the Ovary of Nile Tilapia, Oreochromis niloticus, During Vitellogenesis Suggests Transcriptional Interaction Biol Reprod, May 1, 2003; 68(5): 1545 - 1553. [Abstract] [Full Text] [PDF]

				P. Melamed, M. Koh, P. Preklathan, L. Bei, and C. Hew Multiple Mechanisms for Pitx-1 Transactivation of a Luteinizing Hormone beta Subunit Gene J. Biol. Chem., July 12, 2002; 277(29): 26200 - 26207. [Abstract] [Full Text] [PDF]

				C. Brendel, L. Gelman, and J. Auwerx Multiprotein Bridging Factor-1 (MBF-1) Is a Cofactor for Nuclear Receptors that Regulate Lipid Metabolism Mol. Endocrinol., June 1, 2002; 16(6): 1367 - 1377. [Abstract] [Full Text] [PDF]